The overall aim of this project is to investigate the structure and dynamics of social networks among older adults, and the role of network system-related characteristics in important late-life outcomes, including well- being, cognitive aging, an survival. Previous research suggests that social relationships and the support they provide are important resources to older adults, and associated with better overall health and well-being. They have also been shown to reduce risk of cognitive decline and mortality. In addition, social networks of older adults tend to shrink substantially as they age, as losses due to death and relocation among close relatives and friends take a growing toll on their social network system. However, much less is known about the social resilience of older adults and the network system in which there social ties are embedded. That is, how do the composition and structure of an older adult's social network affect their ability to form new ties as compensation for losses in their network, or conversely, impede their ability to form new ties, and increase social isolation? In addition, it is unknown how features of their social network system, in its interplay with individual-level risk and protective factors, affect the rate of cognitive aging in this population To address these questions, we propose to use a complex-systems approach and develop a series of simulation models to characterize social network structures of an urban population of older non-Hispanic white and African American adults, as well as the changes in these structures over time. We will also use graphical techniques to visualize the network structure and dynamics in this population. Second, we propose to use the results of the network analysis in a series of regression models to prospectively test the association between specific network features, such as the degree of clustering, density, and centrality, and mortality as well as changes in well-being and cognitive outcomes. This research is facilitated by the availability of existing data on a large cohort (N > 6,000) of older adults with detailed information over a period of about 10 years on their social network characteristics and other psychosocial and health variables relevant to the prediction of well-being, cognitive decline and mortality. To our knowledge, this is the first application of a complex system approach to examine important questions related to the role of social network systems in well-being, cognitive aging and survival in late life.